ChatGPT Images 2.0 lands
- OpenAI released ChatGPT Images 2.0, which produces clearer text, structured visuals, and multilingual outputs. - Tests and coverage highlight its ability to create infographics, slides, maps, manga, and legible UI text. - Publications reported the rollout across ChatGPT tiers and its practical use for design and prototyping workflows (venturebeat.com).
OpenAI rolled out ChatGPT Images 2.0 on April 21, saying the new model is available on all ChatGPT plans. (help.openai.com) (openai.com) OpenAI said the model produces more precise edits, keeps details consistent across an image, and runs up to four times faster than the previous system. The company also released the underlying API model as GPT-Image-1.5 for developers. (openai.com 1) (openai.com 2) Image generators have long been good at style and composition but weak at text, charts, and layouts. OpenAI said this version is built to render text more accurately and follow structured instructions closely enough for diagrams, labels, and interface mockups. (openai.com) (developers.openai.com) VentureBeat reported early tests that produced multilingual posters, full infographics, slide-like layouts, maps, manga panels, and user-interface screens with legible text. Its write-up on April 21 described the outputs as substantially cleaner on text-heavy prompts than earlier image models. (venturebeat.com) That shift puts image generation closer to document work that usually needs design software or manual cleanup. OpenAI’s API documentation now frames image generation as a tool for creating and editing visuals from prompts, rather than only making art. (developers.openai.com) (openai.com) The release also extends OpenAI’s push to fold image tools directly into ChatGPT instead of treating them as a separate product. OpenAI’s March 25, 2025 launch of 4o image generation had rolled out first to Plus, Pro, Team, and Free users, with Enterprise and Education access coming later. (openai.com) This time, OpenAI’s release notes said ChatGPT Images 2.0 is available on all plans from day one. The company’s help page, updated April 22, says users can create images from the sidebar Images tool or by prompting in chat. (help.openai.com 1) (help.openai.com 2) OpenAI has also tied the launch to a broader product pattern: multimodal systems that can read, reason, and generate across text and images in one workflow. The company’s separate “thinking with images” materials describe models that inspect photos, zoom into details, and answer visual questions inside ChatGPT. (openai.com) The immediate test for ChatGPT Images 2.0 is whether those cleaner layouts and sharper text hold up outside demos. OpenAI is pitching it as a built-in ChatGPT feature; developers now get the same push through GPT-Image-1.5 in the API. (openai.com 1) (openai.com 2)